Optimization of Kinematics for Birds and UAVs using Evolutionary Algorithms

The aim of this work is to present a multi-objective optimization method to find maximum efficiency kinematics for a flapping wing unmanned aerial vehicle. We restrained our study to rectangular wings with the same profile along the span and to harmonic dihedral motion. It is assumed that the birdlike aerial vehicle (whose span and surface area were fixed respectively to 1m and 0.15m) is in horizontal mechanically balanced motion at fixed speed. We used two flight physics models to describe the vehicle aerodynamic performances, namely DeLaurier’s model, which has been used in many studies dealing with flapping wings, and the model proposed by Dae-Kwan et al. Then, a constrained multi-objective optimization of the propulsive efficiency is performed using a recent evolutionary multi-objective algorithm called ǫ-MOEA. Firstly, we show that feasible solutions (i.e. solutions that fulfil the imposed constraints) can be obtained using Dae-Kwan et al.’s model. Secondly, we highlight that a single objective optimization approach (weighted sum method for example) can also give optimal solutions as good as the multi-objective one which nevertheless offers the advantage of directly generating the set of the best trade-offs. Finally, we show that the DeLaurier’s model does not yield feasible solutions. Keywords—Flight physics, evolutionary algorithm, optimization, Pareto surface.